The extraordinary genetic diversity among globally circulating HIV-1 strains poses a formidable challenge for HIV-1 vaccine design. Therefore, AIDS vaccines that can induce broadly reactive and potent T cell immune responses are urgently needed. In our recent studies, we have generated two synthetic group M consensus env genes (CON6 and CON-S) of all HIV-1 subtypes. The genetic distance between the group M consensus env sequences and any subtype env sequences is only half of those among subtype env sequences to each other (15% vs. 30%). Our preliminary results have shown that both CON6 and CON-S were biologically functional, preserved key antibody binding epitopes and, most importantly, induced T cell responses in three strains of mice that were more cross-reactive than single subtype env immunogens, and were similar in breadth to a subtype A, B and C env polyvalent immunogen. Project 3 will test T cell immune responses induced by a spectrum of centralized HIV-1 genes in mice using a DMA prime-recombinant vaccinia virus boost screening strategy, and comprehensively evaluate the centralized gene approach for development of HIV-1 immunogens that induce potent cross-reactive anti-HIV-1 T cell responses. We will: 1) identify optimal consensus env immunogens that induce broad cross-reactive T cell responses, 2) determine which centralized immunogen designs can induce the broadest T cell immune responses in mice, 3) determine if the group M consensus gag-pol-nef fusion gene immunogens can induce broader T cell immune responses than contemporary subtype (A, B and C) gag-pol-nef fusion gene immunogens, and 4) study the subtype consensus env immunogens from acute infection viruses and other newer generations of centralized genes from Project 1 to identify optimal immunogens for inducing broadly reactive T cell responses. The immunogens that induce optimal T cell responses will be evaluated in non-human primates in Project 4.